- Disable CRC caching
- Disable background images
- Disable previews
Many of those things already failed, this makes it more explicit. And it prevents wasting memory that's necessary for other basic operations.
I think all images should be disabled, really, but a text fallback must be added first...
* PATCHING.md and some instructions for building imgs
* Add Custom Hardware to the table of contents
* Github Markdown formatting changes
* Update BUILDING.md
I'm not too keen on your changes in BUILDING.md. Once a .img/.fw has been flashed, flash/run/monitor will work regardless of the initial format but your addition implies that those don't apply to .img but they do.
I think what should be done instead is move your new "Flashing" section to be right after the "Build everything" section. Maybe rename it "Initial Flashing" or "First Installation" or something along those lines. Then describe how to flash a .img or a .fw for the first time on a new device.
I'm doing a commit just to quickly show you what I have in mind, feel free to revert or entirely rewrite and continue with your vision :)
* Remove redundant changes in README.md
While previous changes in the README.md introduced unnececary info, I do want to keep some links.
One thing is that a link to building/installing should definetly be higher up, as I wasn't sure where to get started with retro-go until I scrolled down the whole README
Another thing is that I added a link to PATCHING in the "Generic ESP32" install instructions as their was already a mention about porting in it
* Rename PATCHING and replace all occurances/refs
Yup, this name makes much more sense
* Fill in blank info and reorganize rg_tool info
This fills in instructions for the esp_tool info.
Also, it made more sense to describe rg_tool in BUILDING.md as it isn't only used for porting
* Consitency with Links and commands
This adds links (to the relevent section) to unlinked text. Also makes the difference between `./rg_tool.py` and `python rg_tool.py` a little more consistant
* Remove Unused Title in README
* Remove redundant spaces and lines
---------
Co-authored-by: ducalex <ducalex007@gmail.com>
The emulator source is unmodified but it is dynamically patched at compile time (by forcing an include file).
Both keyboard and joystick are emulated. For keyboard I looked up the most used keys in games. With feedback I'll be able to tweak that, but the virtual keyboard also works if other keys are needed.
Save states work through the in-game menu but not yet through retro-go's menu.
Launcher images are also missing.
I was under the impression that go-play's original code (from 2018) was public domain or zlib.
But it was in fact GPLv2 so I have to abide by this and rectify the situation!
Which is fine for me, as almost the entire project is licensed that way already :) .
https://github.com/OtherCrashOverride/go-play/issues/31#issuecomment-403679764
Now that partial update is no longer glitchy, there is no reason to ever disable it.
However, the `Update` setting was also indirectly doing one more thing: It imposed a frameskip of 1 when `Full` was selected.
This did prevent stuttering in some games.
But that's an indirect effect, and even I keep forgetting that it's the reason why it can counter-intuitively improve performance.
So I think it's better to handle the frameskip directly instead.
Still can't officially support it because performance in several emulators is pretty bad and I haven't tracked down why yet.
PNG decoding also fails on many files so there's clearly something weird going on...
I'm trying to flesh out the theming documentation and add some helpful methods.
New methods:
rg_gui_get_theme_value: Get an int (color/size/etc) from the theme
rg_gui_get_theme_image: Load an image from the theme
NOTE: This is completely untested, pushing to dev branch to keep working on it later.
It would be important to see how much memory cJSON wastes (by keeping theme_obj around)...
Rel: PR #86
This is a merge of the changes in gwenesis 858ea15.
I have made a very rudimentary keystore to back the state, but it will need a lot more work to be fast and safe...
Also the following things have been changed fom the original:
- lfo_pm_table was removed from save state
- save VRAM instead of emulator_framebuffer (It is the same data, just a different pointer)
For simplicity I reverted to using a python file to define partitions. The csv would've been nice, allowing `idf.py partitiob-table` to work, but it's pointless if we ignore it and generate the table ourselves anyway (due to configurable apps list)...
rg_tool is now using parttool.py to flash apps instead of raw esptool.py.
This removes the need for --offset/--app-offset but adds 2-3 seconds to the command.
We could cache the table if this becomes an issue...
